Home | Databases | WorldLII | Search | Feedback New Zealand Bill of Rights Act Reports

Electoral Act - Rights of Prisoners to Vote (Inconsistent) (Section 12(a)) [1992] NZBORARp 1 (17 November 1992)

Last Updated: 5 January 2019

Rights of Prisoners to Vote: Bill of Rights

McGrath J J 17 November 1992 Department of Justice

Opinion on consistency between NZ Bill of Rights Act and restrictions on prisoners' voting rights

17 November 1992

The Secretary for Justice Department of Justice Private Box 180 WELLINGTON

Attention: Mr W A Moore, Law Reform Division Dear Sir

Rights of Prisoners to Vote: Bill of Rights

1. Thank you for your letter of 19 October 1992 concerning proposed changes to the electoral law. My opinion is sought as to whether the rule in s.42(1)(d) of the Electoral Act 1956 creates a potential problem in terms of compliance with s.12(a) of the New Zealand Bill of Rights Act 1990 ("the Bill of Rights Act"). The question arises in the context of consideration of inclusion of a provision along these lines in a new Electoral Act
   
2. You have kindly supplied copies of an opinion from Mr Gobbi a legal adviser in the Department's Law Reform Division. Mr Gobbi's opinion is that if New Zealand follows the trend of Canadian authorities this rule is inconsistent with s.12(a) of the Bill of Rights and is not, in terms of s.5 of that Act, a justified limitation. Mr Gobbi's opinion also contains suggestions as to the method of remedying the inconsistencies.
   
3. My opinion is divided into three parts. First, I consider whether the rule is a prima facie breach of s.12(a) of the Bill of Rights Act and conclude that is the case. Second, I consider whether the rule is nonetheless a justified limit in terms of s.5 of the Bill of Rights Act. I conclude it is not a justified limit. The rule therefore in my opinion is inconsistent with the Bill of Rights. In the final part of the opinion I examine means of ensuring consistency with s.12(a).
   

I. Is there a prima facie breach of the Bill of Rights?

4. Section 12 of the Bill of Rights Act provides:

"Every New Zealand citizen who is of or over the age of 18 years -

(a) Has the right to vote in genuine periodic elections of members of the House of Representatives, which elections shall be by equal suffrage and by secret ballot;".

Section 42(1) of the Electoral Act provides that a number of persons are disqualified for registration as electors including:

"(d) A person detained in any penal institution pursuant to a conviction:"

5. On the plain words of the two measures, in my opinion, there is a prima facie breach of the Bill of Rights.
   
6. I acknowledge there is a suggestion in a Canadian Case that this is not a prima facie breach. In particular, Monnin C J M, one of three judges sitting in the Manitoba Court of Appeal in Re Badger and Attorney-General of Canada et al v Re Piche et al and Attorney-General of Canada et al (1988) 55 DLR (4th) 177 at 182 said that he would have been "inclined" to conclude that the equivalent provision in the Canada Elections Act did not breach the equivalent right expressed in s.3 of the Canadian Charter of Rights and Freedoms. This was because the right to vote was not an absolute right. Accordingly it could be limited by Parliament. The Chief Justice, however, acknowledged that the majority of the Canadian cases did not support this approach.
   
7. However I consider that the matter is clear beyond argument. There is a prima facie breach of the s.12(a) right.
   

II. Is the breach a justified limit in terms of s.5?

8. Such a prima facie breach of the Bill of Rights of course may nonetheless be a justified limit on the relevant right in terms of s.5. Section 5 provides that the rights and freedoms in the Bill of Rights:

"... may be subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society."

9. There is no difficulty in this case in meeting the first part of s.5, i.e., that the limit is "prescribed by law" assuming that the rule was expressed along similar lines to s.42(1)(d) and was included in statute. In terms of what is meant by that phrase, the New Zealand Court of Appeal has referred favourably to the Canadian approach. See, for example, Ministry of Transport v Noort, Police v Curran [1992] NZCA 51; [1992] 3 NZLR 260 at 272 per Cooke P, 283 per Richardson J and 295 per Gault J. Richardson J noted that the requirement that the limit be prescribed by law ensures that if rights are to be abridged then the abridgements:

"... should be imposed by law so that they are adequately identifiable and accessible by members of the public, and further are formulated with sufficient precision to enable citizens to regulate their conduct and to foresee the consequences which a given action may entail.. .".(at 283)

10. The second part of s.5 raises more difficult questions. As to the limbs of this test, the indications are that the Canadian approach reflected in cases such as R V Oakes (1986) 26 DLR (4th) 200, Re A reference re Public Service Employee Relations Act [1987] 1 SCR 313 and in R v Keeastra [1991] 2 WWR 1 will be followed in New Zealand. Richardson J in Noort approved that and said that it was a matter of weighing:

"(1) the significance in the particular case of the values underlying the Bill of Rights Act;

(2) the importance in the public interest of the intrusion on the particular right protected by the Bill of Rights Act;
(3) the limit sought to be placed on the application of the Act provision in the particular case; and
(4) the effectiveness of the intrusion in protecting the interests put forward to justify those limits." (at p.284)

11. I will apply each limb of that test to the circumstances of this case shortly. Before doing that, I note that the point at issue has been addressed in a number of Canadian cases. I do not however find the Canadian cases of much assistance in advising on the likely approach that would be adopted in New Zealand. First, the Canadian cases have largely been decided at the lower Courts level. The matter has not been addressed by the Supreme Court of Canada. Of the higher level cases the Manitoba Court of Appeal in Re Badger has upheld the limit as a justifiable one whereas the Federal Court of Appeal has reached the opposite view (cf Belczowski V The Queen (1992) 90 DLR (4th) 330). Second, sometimes the compelling nature of the circumstances of particular cases has meant that the relevant issues were not fully canvased. For example the decision in Re Badger seems to have been affected by the short notice given the Court to decide the issue. When the reasons for judgment were released only 80 hours remained before the polls opened for Manitoba's General Federal election. The lower Court decision had required the Chief Electoral Officer to list inmates and to allow those qualified to vote in that election. That decision was reversed. Similarly, in Re Paul et al and Chief Electoral Officer et al: Manitoba Metis Federation, Intervener (1990) 72 DLR (4th) 396 Morse J, of the Manitoba Court of Queen's Bench, who also had been called upon to decide this issue "at the last minute" said that he did not have the right or authority to declare valid legislation which another judge of the same Court had declared to be of no force and effect. While a clear majority of the Canadian cases say the rule in s.42(1)(d) is not a justified limit, for the reasons given little weight should be placed on that. Where the Canadian cases do, however assist is in identifying the relevant issues.
    
12. I now take each limb of the test identified by Richardson J in Noort in turn.

(a) Significance in the particular case of the values underlying the Bill of Rights Act.

13. Obviously, the right to vote is fundamental to a democracy. This point is made in a number of the Canadian cases (e.g. in Badger et al v Attorney-General of Manitoba (1986) 30 DLR (4th) 108 Scollin J. of the Manitoba Court of Queen's Bench described the right to vote as a "keystone right", at 112). The Canadian cases do, however, also acknowledge that there have been limits on this right throughout its history, for example, Monnin C J M and Lyon J A in Re Badger at 182 and 191. Lyon J A in that case noted that the right to vote:

"Has been hedged about with various restrictions, conditions and disqualifications from time to time." (at 191).

14. I therefore proceed on the basis that the values underlying the relevant right are important, seen as fundamental to a democracy but, at times the right has nevertheless been restricted.

(b) Importance in the public interests of the intrusion on the particular right protected by the Bill of Rights

15. I consider that this limb of Richardson J's test requires an examination of the objectives of the limit (and equates with the "pressing and substantial" objective requirement of the Oakes test). As has been noted in relation to the Canadian equivalent to s.42(1)(d), the subsection itself does not on its face give an indication of its purpose. However, in my view, the Canadian cases highlight the arguable objectives which can be advanced in relation to the restriction on the right of inmates to vote.
    
16. First, one possible objective is that practical considerations support the limit. The British Colombia Supreme Court in Re Jolivet and Barker and The queen and Solicitor- General of Canada (1983) 1 DLR (4th) 604 concluded that if the right to vote comprised simply the right to mark a ballot paper and to have it counted then it could be assured to prison inmates with no great difficulty. Taylor J concluded, however, that the right to vote means more than the right to case a ballot:

"It means the right to make an informed electoral choice reached through freedom of belief, conscience, opinion, expression, association and assembly. That is to say with complete freedom of access to the process of 'discussion and the interplay of ideas' by which public opinion is formed." (at 607)

17. The Judge concluded that the exercise of the vote in the circumstances of restrictions imposed by imprisonment on the freedom of the person made it impossible for prisoners to make a free democratic electoral choice. Casting a ballot in such circumstances could not be properly described as an exercise of the right to vote. Denial of the right was a matter of necessity thus, the Court found that the relevant provision in the Canada Elections Act was a justifiable limit on the right under s.3 of the Canadian Charter. By contrast this same objective was rejected as the basis of a valid limit by the Federal Court trial division in Levesque v Attorney-General of Canada et al (1985) 25 DLR (4th) 184. Rouleau J's view was that the fact that some of the rights of prisoners were necessarily curtailed for administrative or security reasons did not justify curtailing the whole spectrum of rights (at p.189). That other State jurisdictions were able to accommodate the right was important.
    
18. In terms of the New Zealand system, I am not aware of any administrative or security difficulties resulting from the exercise of the right to vote by prison inmates which would constitute an objective of sufficient importance in the public interest in terms of the Noort test. In those circumstances this factor is not compelling.
    
19. The second possible objective identified by the Canadian cases is that deprival of the right to vote should be part of the sanction for those who have committed offences. This objective was accepted as an appropriate one in terms of the Oakes test in both Badger cases. Lyon J A in Re Badger noted that when persons were imprisoned they were deprived of a number of rights and freedoms and that one of those rights was the right to vote (at 193). He saw the relevant provision in the Canada Elections Act as merely confirming that consequence. This objective was accepted as "more plausible"

than the other objectives advanced at first instance in Belczowski where Strayer J. noted that the relevant provision did not disqualify those who were in prison awaiting trial or those charged with offences but not convicted (at 236). Moreover deprivation of the right only lasted as long as imprisonment. Strayer J's approach on this point was however rejected on appeal by the Federal Court of Appeal where the Court said that this and the other objectives advanced by the appellant were merely symbolic and abstract and as such could not be regarded as "pressing and substantial" (at 340-341).

20. I consider that this objective is a legitimate one and that it can be of sufficient importance in the public interest to justify some intrusion on the right to vote. Inmates' other rights such as the right to freedom of assembly and association are abridged on the basis that this is part of the sanction for committing offences. I see some infringement on the right to vote as also a valid objective in this context, subject to the effect not being arbitrary, a matter which I discuss later.
    
21. The third objective that has been advanced in the Canadian context is that of maintaining the security of the franchise by ensuring its exercise by a decent and responsible citizenry. This was accepted as a valid objective in Sauvé v Attorney- General of Canada et al (1988) 53 DLR (4th) 595 a decision of the Ontario High Court of Justice. In that case the Court had heard evidence that a vote was deemed to be more likely to be responsible if the person casting it had a demonstrable stake in the community and its public affairs, took an active interest in public affairs, and was adequately informed about public issues (at 599). This approach was also accepted by the Manitoba Court of Queen's Bench in Badger but was rejected at both levels in the Belczowski case and also in Re Grondin v Attorney-General of Ontario et al (1985) 65 OR (2d) 427 a decision of the High Court of Justice of Ontario. In Belczowski the Federal Court of Appeal noted that the fact of being in prison is not by any means a sure or rational indication that the prisoner "is not a decent and responsible citizen." Furthermore there were many indecent and irresponsible persons outside of prison who were able to vote (at 343).
    
22. I do not consider that any significant weight can be given to this objective in terms of the public interest.
    
23. The final objective identified in the Canadian cases is that of preserving the integrity of the voting process. The idea is that as voting is more than marking the ballot paper it is difficult to exercise this right properly in prison. This approach was rejected, correctly in my view, at both levels in the Belczowski case. The Court noted that the inmate bringing the action had been able to follow events on television to some extent while in prison. That would also be the case in New Zealand. In any event there would be other persons, for example those in isolated places, who on this approach would be disenfranchised.

(c) Limits sought to be placed in the particular case

24. There is a difficulty in applying this part of the test because the limit is not confined in any way. Indeed, as Strayer J. said at first instance in Belczowski, it is a "direct frontal assault" on the right of inmates to vote (at 237).

(d) Effectiveness of the intrusion in protecting the interest put forward to justify those limits

25. This limb of the Noort test equates with the requirement in Oakes that there be some rational connection between the limit and the objective and some proportionality. The Canadian cases identify the difficulty with such a limit in relation to inmates' voting rights. First, the limit is somewhat arbitrary as it applies to any type of offence no matter how serious. Second, its actual operation can mean that a person's rights are determined by fortuitous circumstances. For example1 a person convicted of a minor offence may, because of timing, lose the right to vote for three years whereas someone convicted of a more serious offence may not lose the right at all. Even if all of the objectives identified above were appropriate in terms of the public interest, in my view this part of the test would not be met.
    

III. Means of ensuring compliance

26. The major criticism identified by the Canadian cases of the provision relates to its arbitrary effect. That is because he nature of the offence committed is irrelevant in terms of the effect on voting rights. Accordingly, I agree with Mr Gobbi that it would be necessary to look at an approach along the lines of that adopted in Australia and make some differentiation on the basis of the seriousness of the offence to avoid this problem. In this regard, I see some merit in the approach recommended by the Royal Commission on electoral law. That is, that prisoners who following conviction have been sentenced to a term of three years or more should not be allowed to vote. (Recommendation 42). I appreciate that the Royal Commission reported prior to passing of the Bill of Rights Act. As a result it can no longer be argued that voting is merely a privilege. However the logic of the Royal Commission's view is still persuasive. I would regard this approach as one which can minimise the problem of arbitrary application and is valid in terms of the second objective discussed above.
    

Yours faithfully

J J McGrath QC Solicitor-General